Bilingualism and cognition: The impact of age of acquisition, language use and proficiency in healthy older adults

Background The positive effect of bilingualism, especially later in life, is typically attributed to its role as a cognitive reserve factor. However, evidence supporting such benefits remains inconclusive or appears limited to certain populations, tasks, or cognitive domains. Some of these inconsistencies may stem from the tendency to define bilingualism as a dichotomous variable, rather than considering the multiple factors that contribute to the bilingual experience. Our study investigates the impact of bilingualism on cognition in cognitively unimpaired older adults, focusing on age of language acquisition (AoA), proficiency, and usage throughout life. Method We analyzed data from 2415 cognitively unimpaired individuals (aged 45–74) from the Alzheimer's and Families (ALFA) study. We included Mini-Mental State Examination to assess global cognition, semantic fluency for lexical retrieval, Memory Binding Test for verbal episodic memory, and WAIS-IV subtests for processing speed and visual-motor coordination (Coding), visual-spatial reasoning (Visual Puzzles), non-verbal abstract reasoning (Matrix Reasoning), verbal short-term memory (Digit Span Forward) and working memory and attention (Digit Span Backward and Sequencing). We defined three groups based on AoA (Early/Late) and proficiency (High/Low) of Catalan: 1) Early High-Proficiency bilinguals (n = 1559); 2) Late High-Proficiency bilinguals (n = 537); and 3) Late Low-Proficiency bilinguals, primarily Spanish-dominant (n = 319). We also analyzed the effect of the language of assessment (Spanish or Catalan) on cognitive performance. Results We found that both Early and Late High-Proficiency bilingual groups outperformed Late Low-Proficiency bilinguals in verbal fluency and processing speed. Additionally, Early High-Proficiency bilinguals scored significantly higher in verbal short-term memory than both Late AoA groups. Conclusion Our findings highlight that the cognitive effects of bilingualism in aging are not uniform but rather domain-specific and that bilingualism, particularly when characterized by high proficiency, can serve as a meaningful contributor to cognitive reserve.